Growth mechanics work. That's the uncomfortable part about them — they're not just annoying, they're effective, and that's exactly why leaving them out takes a deliberate decision rather than just forgetting about them. Here's the list of what we've left out on purpose, with the reasoning behind it.
The 9pm push
We had it briefly, in an internal test build: an evening reminder, timed to when usage numbers are, by experience, easiest to lift, because most people have time and a phone in hand then. It worked — return rate rose measurably. It also led to half the test team muting it after four days. "The 9pm pushes were hell," read our internal feedback channel, verbatim. When your own team, who knows exactly why the notification is arriving, still wants to get rid of it, the answer is clear. Out it went.
"We miss you"
The classic email after seven or fourteen days of inactivity, usually dressed up with a discount or a fake update notice to manufacture a reason to come back. The pattern behind it: absence gets treated as a problem to solve — not as what it often simply is, namely that there was nothing to say right now. We don't want an app to make you feel like you owe it something just because you weren't in it for a few days.
No newsletter with statistics about you
A popular variant: a weekly summary — "You saw 12 posts this week, 3 people reacted to you" — whose main purpose is to remind you of the app by simulating activity you'd long since forgotten. Building a summary like that would also require us to log what you see and how often — data we don't want to collect in the first place, for exactly this reason. No newsletter, because it would require a surveillance apparatus we don't have.
Streaks and counters that can break
Daily streaks — "5 days active in a row!" — trigger one of the most effective psychological levers there is: the fear of losing something you've built up. That's exactly why they're so widespread, and exactly why we don't build them. A streak rewards presence, not content. Whoever only checks in because of the number has nothing to say — and we wanted an app people check into because they want to say something, not to keep a number alive.
"X people also saw this"
Social-proof cues — showing how many others are doing or seeing the same thing right now — create urgency where there objectively isn't any. They work because people orient themselves by the crowd, not because the information itself is useful. We show you what's happening around you, but not as a counter meant to push you into reacting faster.
What's left once you drop all of that
Honestly: less. Usage numbers would be higher with every one of these mechanics, we won't pretend otherwise — some of them we tested and pulled out again for exactly that reason, not because they didn't work, but because they worked too well for what they actually did. What's left is an app that doesn't try to pull back anyone who wouldn't come back on their own. That's slower growth. It's also the only measure we really want to be judged by: whether someone stays because they want to, not because we made them.